The Agent Skills Directory

[SAFE]: The skill defines a set of API endpoints for TikHub to interact with TikTok data, such as fetching profiles, videos, and trends.
[DATA_EXPOSURE]: The skill facilitates network requests to api.tikhub.io. This is identified as a legitimate vendor resource for TikHub and is consistent with the skill's stated purpose.
[CREDENTIALS_UNSAFE]: The skill uses the environment variable TIKHUB_API_KEY for authentication. This is a recommended security practice for managing secrets in AI agent skills.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data such as video URLs, user IDs, and search keywords.
Ingestion points: Inputs for parameters like share_url, keyword, and unique_id in SKILL.md.
Boundary markers: None identified in the provided documentation.
Capability inventory: Network requests via curl are documented as the primary capability.
Sanitization: No explicit sanitization or validation of the input parameters is described in the instructions.

tiktok